Latest Articles for Speech Recognition

A new method improves speech act recognition in Bengali using audio and text analysis.

A new approach improves speech language identification using self-supervised learning and labels.

A new method enhances speech recognition for dysarthric Arabic speakers.

Allophant enhances phoneme recognition for languages with limited data.

Improving how speech recognition systems estimate word timing for better accuracy.

New methods enhance speech processing in language models.

Discover a new method for combining different types of data effectively.

Self-supervised models reveal insights into phonetic and phonemic distinctions in speech.

Research reveals how our brain tracks speech features during comprehension.

This study focuses on improving spoken NER through transfer learning and E2E models.

A new method enhances task-oriented dialogue systems using audio and knowledge integration.

Recent research improves ASR models for Norwegian, enhancing performance in Bokmål and Nynorsk.

New methods improve multilingual speech recognition using existing data sources.

Research focuses on enhancing speech tech for languages lacking sufficient data.

This article discusses a new method for building efficient ASR systems.

CML-TTS enables better text-to-speech systems across seven languages.

SURT 2.0 improves speech recognition for multiple speakers in real-time settings.

A new method enhances speech recognition technology without losing previously learned knowledge.

A new method evaluates ASR systems without needing reference texts.

NoRefER offers a new way to assess speech recognition outputs without needing transcripts.

New methods enhance speech segmentation in multi-language conversations.

A new framework improves ASR for low-resource languages and multilingual scalability.

A new method enhances lip reading accuracy using visemes in speech recognition.

Personalized ASR systems improve communication for DHH individuals significantly.

New methods leverage conversational summaries for better speaker recognition.

Enhancing feedback systems for English learners by addressing the cold start problem.

A new model enhances speech extraction using audio and visual information.

Learn how new techniques improve speech clarity in noisy environments.

This article discusses new models that enhance speech recognition accuracy by considering longer context.

New method enhances learning in Spiking Neural Networks by incorporating delay adjustments.

Research highlights methods to protect gender privacy in spoken audio.

New framework enhances speech clarity from silent videos through improved processing.

Researchers develop a Conformer model to improve fake audio detection.

Research on improving acoustic word embeddings with semantic understanding and multilingual data.

A new approach that combines speech with language models for improved translation.

New methods enhance speech recognition accuracy, addressing common transcription errors.

This article explores a new model for speech intent and slot identification.

New method improves speech recognition using only raw audio data.

A study enhances ASR for older speakers, using innovative techniques.

ivrit.ai provides vital resources for enhancing Hebrew ASR technology.